**Background:** Pterygium is a common ocular surface disease characterized by abnormal conjunctival growth onto the cornea, with prevalence ranging from 0.07% to 53% across populations. While environmental triggers like UV exposure are known, the role of circulating lipids remains controversial. Observational studies have reported conflicting associations between lipid levels and pterygium, but these are prone to confounding. Mendelian randomization (MR) uses genetic variants as instrumental variables to infer causal relationships, overcoming confounding and reverse causation. This study aimed to assess the causal effects of circulating lipids (HDL-c, LDL-c, TC, TG) on pterygium risk and to evaluate the potential of lipid-lowering drugs (HMGCR and PCSK9 inhibitors) as therapeutic targets.
**Methods:** A two-step MR design was employed. In step 1, bidirectional two-sample MR was performed using summary statistics from independent GWAS of European ancestry: HDL-c (n=403,943), LDL-c (n=201,678), TC (n=94,595), TG (n=78,700), and pterygium (n=203,880; 363 cases, 203,517 controls). SNPs associated with each lipid trait at p<5×10⁻⁸ were selected as instruments after linkage disequilibrium clumping (r²<0.001, window 1 Mb). Inverse variance weighted (IVW) was the primary analysis, with MR-Egger, weighted median, weighted mode, and simple mode as sensitivity analyses. In step 2, drug-target MR used cis-eQTLs (within 1 Mb of target gene) from the eQTLGen Consortium for HMGCR (921 cis-eQTLs) and PCSK9 (24 cis-eQTLs) as instruments. These were further filtered to SNPs associated with LDL-c (p<5×10⁻⁸) and low LD (r²<0.30), yielding 7 SNPs for HMGCR and 15 for PCSK9. Coronary artery disease (CAD) was used as a positive control to validate instrument strength.
**Key Results:** Forward MR showed a significant positive association between genetically predicted LDL-c and pterygium risk (IVW OR=2.227; 95% CI: 1.15–2.33; p=1.53×10⁻⁴). Results were consistent across MR-Egger (OR=3.261; p=1.71×10⁻⁴), weighted median (OR=2.381; p=4.17×10⁻³), and weighted mode (OR=2.412; p=2.21×10⁻³). Similarly, TC was positively associated with pterygium (IVW OR=1.806; 95% CI: 1.248–2.613; p=1.70×10⁻³), with consistent estimates from MR-Egger (OR=2.238; p=0.010), weighted median (OR=2.213; p=2.81×10⁻³), and weighted mode (OR=2.302; p=1.99×10⁻³). No significant associations were found for HDL-c (IVW OR=1.025; p=0.891) or TG (IVW OR=1.160; p=0.485). Reverse MR indicated no causal effect of pterygium on any lipid trait (all p>0.05). Drug-target MR revealed a significant positive association between genetically proxied HMGCR expression (i.e., higher HMGCR activity, which increases LDL-c) and pterygium risk (IVW OR=6.999; 95% CI: 1.436–34.114; p=0.016), suggesting that HMGCR inhibitors (statins) may reduce pterygium risk. In contrast, PCSK9 expression showed no significant association with pterygium (IVW OR=1.479; p=0.322). Positive control analyses confirmed that both HMGCR and PCSK9 instruments were strongly associated with CAD (HMGCR: IVW OR=1.569, p=1.45×10⁻⁵; PCSK9: IVW OR=2.211, p=2.54×10⁻¹⁷).
**Clinical Implications:** This study provides genetic evidence that elevated LDL-c and TC are causal risk factors for pterygium, supporting the potential of cholesterol-lowering strategies for prevention. The drug-target MR specifically suggests that statins (HMGCR inhibitors) may be effective in reducing pterygium risk, possibly through anti-inflammatory and antioxidant mechanisms beyond lipid lowering. PCSK9 inhibitors did not show a similar effect, indicating that the protective effect may be specific to HMGCR inhibition. These findings warrant further clinical trials to evaluate statin therapy as a preventive or adjunctive treatment for pterygium, particularly in individuals with hyperlipidemia. Limitations include the European-only ancestry, modest sample size for pterygium (363 cases), and the need for mechanistic studies to clarify the role of cholesterol metabolism in pterygium pathogenesis.